I’d like to point you towards several articles around the ongoing Oracle bun-fights I mentioned in a previous posting. I don’t think I should add any comment other than I hope lots of people get to read these different thoughts and decide for themselves who they want to trust.
First, what Tom Kyte suggested could be considered a trilogy (only two parts so far, but I’m waiting enthusiastically for the third)
Can we have a sensible debate ?
Tom Kyte Thread on ‘Proof’
Next, some of the threads that inspired these responses.
Understand the Oracle �proof??? Myth
Do single-user scripts “prove” Oracle Performance?
Unfortunately, the latter seems to have been edited during the time I’ve been following it, but you should still manage to get the gist.